To help settle this, I'll define color in such a way as to divorce it from perception by eyes:

color n. A property of electromagnetic radiation such that one band of wavelengths is more intense than an adjacent band.

© Yerrick 2001.

Red objects are perceived as red because the cones of our eyes respond to wavelengths around 700 nm differently from 580 nm (yellow), 520 nm (green), or 470 nm (blue). ariels claimed that the color magenta was a counter-example, but it merely consists of power peaks in both the visible blue and visible red bands.

This definition of "color" also generalizes to radio transmission and receiving (radio stations come in on different "colors") and reminds a fellow of how fiber optics works. Salon.com has published an article (http://www.salon.com/tech/feature/2003/03/12/spectrum/index.html?x) about how this conception of spectrum as color could make the FCC's job less important.